2016 elections - because it's never too early

[The Brain]

The whole district should be held responsible if one of its inhabitants break the law. Also passports should be required for domestic travel.

[Tonitrus]

The whole district should be held responsible if one of its inhabitants break the law. Also passports should be required for domestic travel.

No papers! 

[Eddie Teach]

No papers! 

That doesn't look like Indiana Jones. 

[Razgovory]

The argument from geography is that some people's  votes should mean less then others, and is an overt endorsement of gerrymandering.

Not if you believe that districts should be geographically compact--and districts that aren't geographically compact are basically the most obvious sign of gerrymandering.

To draw the districts the way you seem to want--so that each district would not only be roughly equal in population, but would also have roughly the same overall party balance as the state as a whole--you'd probably have to throw out trying to have geographically compact districts in most states.  Instead of trying to simply divide the state into compact districts with equal population--what neutral redistricting commissions are supposed to try to do--you'd have to explicitly consider demographic factors and past voting patterns in drawing up the districts--which is what you do when you gerrymander.

And even then, it probably wouldn't give the result you want.  If the state as a whole voted 51/49 for a particular party, and each district in the state reflected the demographics of the state as a whole, then the same party would often win every district in the state by a 51/49 margin.

I think the real problem is that you don't realize just how heavily towards one party or the other many areas of the country skew.  The fact is, while in the country as a whole, the 2 parties have roughly equal support, when you start looking at smaller areas, in many of them one of them has overwhelming support.  I remember looking at some maps showing voting results after the 2000 Presidential election.  Of course, that election was very close in the popular vote, but there were in fact very few Floridas--states where the popular vote was close.  And when you started to look at county-by-county vote totals, the results were even more striking.  Even in competitive states like Florida, there were very few counties where the vote close.  There were hardly any counties anywhere that had anything close to a 50/50 split in the popular vote.  Most counties went quite heavily one way or the other.  Counties where one party got at least 60% of the vote were very common;  in many counties the splits were 70/30, 75/25, or even more lopsided.

I haven't looked at more recent election data in detail, but I very much doubt that there's been much overall change.

Question:  Can they be compact and not provide such bizarre results as more people voting for a party yet that party see dramatically reversal?

[Valmy]

People living in cities have less of a right to representation. Everybody knows this.

[DGuller]

Question:  Can they be compact and not provide such bizarre results as more people voting for a party yet that party see dramatically reversal?

It seems very unlikely.  And I'm hedging my words only because I've never dug deep into the calculations, so I may be unaware of some subtleties in this particular case.  If your criteria is geographical compactness, there isn't really much freedom there in defining your clusters, it's a rather simple mathematical algorithm.

In way we're trying to square a circle here.  If you we want to have a system where every vote counts equally, then the only way to go is having party lists.

[Razgovory]

Question:  Can they be compact and not provide such bizarre results as more people voting for a party yet that party see dramatically reversal?

It seems very unlikely.  And I'm hedging my words only because I've never dug deep into the calculations, so I may be unaware of some subtleties in this particular case.  If your criteria is geographical compactness, there isn't really much freedom there in defining your clusters, it's a rather simple mathematical algorithm.

In way we're trying to square a circle here.  If you we want to have a system where every vote counts equally, then the only way to go is having party lists.

Well, if one area is overly compact (like a city), then it stands to reason that other areas will have to be less compact as they will need to be larger.  The second question is "are the districts compact?"



Someone went to a lot of trouble to create that one democratic district in North Carolina.  I would say, that no, the redistricting is not particularly compact.  A lot of what I see is one district in a city and a lot of surrounding built around siphoning away democratic votes.

[alfred russel]

It seems very unlikely.  And I'm hedging my words only because I've never dug deep into the calculations, so I may be unaware of some subtleties in this particular case.  If your criteria is geographical compactness, there isn't really much freedom there in defining your clusters, it's a rather simple mathematical algorithm.

In way we're trying to square a circle here.  If you we want to have a system where every vote counts equally, then the only way to go is having party lists.

If we take the example of Georgia, the state is roughly 55-45% republican, with the Democrats closing fast.

The issue is half the people live in the metro Atlanta area, and while it is very vaguely defined, lets say half are in the city and half in he suburbs.

The actual inner city of Atlanta is massively democratic--republicans are complete non entities. The suburbs are probably closer to 50-50 depending on where we are talking about--maybe a slight lean republican. The rest of the state trends republican, but not anywhere like what is in inner city Atlanta.

So if you start drawing districts by grouping the most densely populated communities as a unit, you end up creating massively democratic majority districts in Atlanta.

What is interesting is that the Republicans in Georgia have gotten very aggressive about maximizing the number of republicans in government. So while there are a few democratic districts with massive majorities, in many cases the strategy has been to create lots of republican districts with very shallow majorities. If you think about this in theory: if the republicans create all districts so that they mirror the voting patterns of the larger state, they will sweep all of the seats so long as they keep a majority. The democrats are openly talking about how this is going to end up as their advantage: if the state ends up mildly democratic in an election cycle--which with the demographic changes may be soon--they can end up sweeping a very large number of seats.

[DGuller]

Well, if one area is overly compact (like a city), then it stands to reason that other areas will have to be less compact as they will need to be larger.

That's not what compact means.  Compact doesn't mean dense, it means that some measure of distance between all points is minimized.  It doesn't have to be small, just as small as possible.

[dps]

Someone went to a lot of trouble to create that one democratic district in North Carolina.  I would say, that no, the redistricting is not particularly compact.  A lot of what I see is one district in a city and a lot of surrounding built around siphoning away democratic votes.

I think it was pointed out somewhere upthread that NC is one of the most heavily gerrymandered states, so good work cherry-picking it.  OTOH, it was also pointed out that some of those obviously non-compact districts were created not to disenfranchise Democratic voters as a whole, but to create districts in which minorities (read:  blacks) are a majority of the voters.  So the effect of a couple of those weird districts in NC is to create safe districts for Democrats (because a heavily black district is effectively a safe Democratic district).

[Razgovory]

Well, if one area is overly compact (like a city), then it stands to reason that other areas will have to be less compact as they will need to be larger.

That's not what compact means.  Compact doesn't mean dense, it means that some measure of distance between all points is minimized.  It doesn't have to be small, just as small as possible.

No but the more "compact" a district is, the more densely populated will be.  Anyway, that's really irrelevant now.  The map I posted that demonstrated the districts not particularly "compact", some are coiled around like a snake and gives lie to the argument that situation is an artifact of honest attempts at geography.  With such bizarre shapes created by redistricting, I think it's clear that geometry or geography was less a concern then electing one particularly party even if it contradicts the will of the electorate.

[Razgovory]

Someone went to a lot of trouble to create that one democratic district in North Carolina.  I would say, that no, the redistricting is not particularly compact.  A lot of what I see is one district in a city and a lot of surrounding built around siphoning away democratic votes.

I think it was pointed out somewhere upthread that NC is one of the most heavily gerrymandered states, so good work cherry-picking it.  OTOH, it was also pointed out that some of those obviously non-compact districts were created not to disenfranchise Democratic voters as a whole, but to create districts in which minorities (read:  blacks) are a majority of the voters.  So the effect of a couple of those weird districts in NC is to create safe districts for Democrats (because a heavily black district is effectively a safe Democratic district).

I think you misunderstand the process of Gerrymandering.  You create one or two very safe districts of Democrats and then dilute the rest of the democrats in the state by mixing in a comfortable majority of rural Republicans.  Such a hypothetical situation would create districts like  seen in Ohio and Indiana. I also see pretty bold attempts at gerrymandering in Michigan, Texas, Pennsylvania, Wisconsin, Minnesota, Florida, etc.  Just look at all the elongated red districts on the map.  Are you honestly going to tell me this is what passes for compact?  A long thin district in Virginia the borders Maryland and South Carolina?  If it was compact you'd expect the shape to be closer to solid square block.  Most of these are anything but.

[dps]

Well, if one area is overly compact (like a city), then it stands to reason that other areas will have to be less compact as they will need to be larger.

That's not what compact means.  Compact doesn't mean dense, it means that some measure of distance between all points is minimized.  It doesn't have to be small, just as small as possible.

No but the more "compact" a district is, the more densely populated will be.  Anyway, that's really irrelevant now.  The map I posted that demonstrated the districts not particularly "compact", some are coiled around like a snake and gives lie to the argument that situation is an artifact of honest attempts at geography.  With such bizarre shapes created by redistricting, I think it's clear that geometry or geography was less a concern then electing one particularly party even if it contradicts the will of the electorate.

I don't think that anybody was contradicting that, exactly.  What we're saying is that 1) districts like what you see in NC are now outliers, whereas they used to be the norm, and 2) that gerrymandering creates safe districts for one party or the other, but it probably doesn't have that much of an overall effect on the total numbers of Representatives elected from each party--it just makes the primary in a safe district the de facto general election.

[Razgovory]

Oh?  What evidence do you have that?

[dps]

Oh?  What evidence do you have that?

Of which?